Although diabetes is a lifelong, chronic disease, the majority of intervention studies to enhance and support diabetes self-management have focused on short-term interventions. Short-term diabetes self-management interventions without follow-up have only short-term (generally less than six months) positive effects on self- management behavior and glycemic control. This study will compare the effectiveness of access to a weekly community-based group (CBG) intervention, and an enhanced usual-care (EUC) control group in helping urban African-American adults with type 2 diabetes achieve and sustain effective self-management, quality of life, and diabetes-related health gains over the long term (three years duration). Specific Aim 1: To conduct a longitudinal randomized controlled trial to evaluate the relative effectiveness of access to a weekly community-based group (CBG) intervention, and an enhanced usual care (EUC) control group. Specific Aim 2: To determine predictors (attendance patterns, demographic, behavioral, health-related) of patients who benefit the most from access to a weekly CBG intervention and EUC control group. Specific Aim 3: To assess the cost, health effects, and cost-effectiveness of access to the CBG intervention versus EUC control group in the short term. Specific Aim 4: To estimate the expected effect that access to the CBG intervention and EUC control group have on end-stage patient outcomes (both microvascular and macrovascular complications) and long-term costs through the use of Monte Carlo simulation models. Hypotheses 1: Access to the CBG intervention will prove more effective than the EUC control group in helping patients achieve and/or sustain improvement in A1C and diabetes-related quality of life. Hypotheses 2: Access to the CBG intervention will be more cost-effective than the EUC control group. The cost of diabetes in the United States has reached $132 billion per year accounting for 10% of total health care expenditures. These costs are expected to increase steadily over the next 20 years. African Americans are two times more likely to have diabetes than Caucasian Americans. The purpose of this study is to develop and evaluate a community-based program to help African Americans with diabetes prevent complications, thus reducing the financial and human cost of this illness.